Automate All the Things Panel Discussion
summary
Conference cites 17% of teams using AI-driven cloud ops and 41% planning adoption within six months while none are fully AI-managed yet, and the panel argues concrete wins come from human-in-the-loop workflows not full autonomy. Demonstrated use cases include autogenerating Terraform/Helm, RAG-enhanced reasoning, agent-driven refactors that shrink files from 850 to 180 lines but introduce subtle bugs, a push toward hosting open-source models on commodity GPUs, and treating LLMs plus curated RAG data as complementary engineering tools.
AI-written summary · about
chapters (11)
- 0:00Introduction to AI in IaC
- 1:30Panelist introductions
- 3:59AI-generated infrastructure code
- 7:52AI ops vs MLOps debate
- 12:25RAG data vs LLM importance
- 15:11Adapting engineering skills for AI
- 20:53Can AI fully automate IaC?
- 21:15Homegrown vs third-party AI models
- 24:32AI capability peak concerns
- 27:03AI-proofing tech interviews
- 29:09Closing remarks and thanks
related talks
41mPanel discussion: AI in code development – hype vs reality
44mPannel Discussion | Agentic Automation: Are AI Agents The Future of Digital Work
46mLeveraging AI and Automation in DevOps
52mPanel: Productivity gains through AI?!
39mDevOps in the Age of AI: Human Powered Evolution
24mAI Governance for developers: A practical framework for deploying autonomous coding agents safely…
05mAutomating Cloud-Native Infra with Intelligent Tools
26mLeadership in Times of AI